NVIDIA Warp has reached 10 million downloads, driven by the developers and researchers using it to experiment, contribute, and build GPU-accelerated workflows.
Warp is an open-source Python framework and CUDA-X library for authoring high-performance GPU kernels and differentiable workflows in Python. It supports interoperability across PyTorch and JAX, with use cases spanning computational engineering, simulation, geometry processing, robotics, AI, and optimization.
To learn more, join next week’s OpenUSD livestream, Accelerate Computational Engineering and Robotics With NVIDIA Warp, for a closer look at what’s possible with Warp.
Add to Calendar: Link